Google电子表格错误文本结果的文本结果长于50000个字符的极限
我试图将细胞组合并在一个细胞中显示,因为每个细胞都包含分开的产品skus逗号。需要在同一列中将这些细胞与分离池中的逗号分离器相结合。 为此,我正在使用,
=TEXTJOIN(",",TRUE, G5,G10,G19,G27,G39,G46,G59)
但会出现错误:
文本Join的文本结果比50000个字符的极限长。
I am trying to combine cells and show in one cell as each cell contains product skus comma seperated. Need to combine these cells with comma seperator in seperate cell in same column.
For this i am using
=TEXTJOIN(",",TRUE, G5,G10,G19,G27,G39,G46,G59)
But getting error:
Text result of TEXTJOIN is longer than the limit of 50000 characters.
如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。

绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱,如果其他用户或者作者回复了您的评论,将不能在第一时间通知您!
发布评论
评论(3)
使用查询(这是唯一的方法):
或:
use query (that's the only way):
or:
根据 @player0的答案以及他对“查询在每个单元格之间添加一个空白空间”的评论,您可以使用
arrayformula
,split
替换函数以操纵查询
函数的输出。= arrayformula(split(split(QUERY({g5; g10; g19; g19; g27; g39; g46; g46; g59}&“@@@@@@@@@@@@@@@9^9),“ @@@@@@@” ,“”),“ @@@@”))
通过设置一个独特的可能(在数据中存在较低的机会)字符,例如“ @@@@@@”,然后在之后添加空间它(例如,“ @@@@”),您可以代替空价值(例如,“”)。但是,最后,它将保留一个“ @@@@@”,而无需空间。为了解决这个问题,您可以将其拆分并仅获取第一部分。
该公式还将达到替代限制,但是它可以维持50000多个字符。
According to @player0's answer and his comment about "query adds one empty space between each cell", you can use the
ARRAYFORMULA
,SPLIT
, andSUBSTITUTE
functions to manipulate the output of theQUERY
function.=ARRAYFORMULA(SPLIT(SUBSTITUTE(QUERY({G5;G10;G19;G27;G39;G46;G59}&"@@@@@",,9^9),"@@@@@ ",""),"@@@@@"))
By setting a unique possible (with lower chance to exists in your data) character, such as "@@@@@", and adding the space after it (e.g., "@@@@@ "), you can substitute for an empty value (e.g., ""). However, in the end, it will keep one "@@@@@" without space. To solve this, you can split it and get the first part only.
This formula will also achieve a SUBSTITUTE Limitation, however, it can sustain a little more than 50000 characters.
板单元不得超过50,000个字符:
我建议您将数据分为几个单元格。
参考:
A Sheets cell cannot have more than 50,000 characters:
I'd suggest you to split your data into several cells.
Reference: